Using electricity customer profiles to combat GHG emissions: New evidence from ComEd AMI data

Abstract

This first-of-its kind study categorizes the residential customer base of Illinois’ largest electric utility, Commonwealth Edison, into slices of customer-usage profiles to determine their marginal Greenhouse Gas emissions for the period 2016−2018. The analysis utilizes anonymous energy-usage data from Advanced Metering Infrastructure, PJM marginal emissions data, and demographic data from the U.S. Census Bureau’s American Community Survey of 2017. Using a machine-learning algorithm called k-means clustering, the analysis identifies distinct usage patterns of residential customers and once again proves the value of access to anonymous AMI data. Even more importantly, the results provide new evidence to inform regulators, consumer advocates, policymakers and utilities on how best to customize energy efficiency, weatherization, customer education, and demand response programs for maximum benefit.